US 7903759 B2 Abstract The invention relates to signal transmitting engineering. The use of the inventive method in systems for transmitting and receiving quadrature amplitude-modulation signals (QAM) exhibiting a low carrier frequency synchronization threshold makes it possible to decrease a demodulation threshold by means of said low carrier frequency synchronization threshold. The result is attainable by supplementing a burst of M m-level sensitive QAM symbols by predetermined symbols, one part of which remains constant form one burst to another and the other part is periodically invertible in certain bursts, thereby the QAM signal components corresponding to the additional predetermined symbols (whose frequencies are known) are extracted on a receiving side. The inversion frequency is determined according to said components, thereby making it possible to eliminate the ambiguity of the receiving frequency synchronization control and to approach the Shannon's threshold.
Claims(20) 1. A method for transmitting and receiving quadrature amplitude modulation (QAM) signals, the method including steps of:
at the transmitting side:
a) dividing the information sequence of m-level symbols, where m =2
^{k}, k =1, 2, 3 . . . running with the first clock frequency f_{1}, into an I channel of the transmitting side with the even m-level symbols and Q channel of the transmitting side with the odd m-level symbols, the m-level symbols in each of the I and Q cannels of the transmitting side running with the frequency f_{1}/2;b) in each of the I and Q channels of the transmitting side, forming bursts of M m-level symbols at the interval
where 2
^{L−1}<M <2^{L}, L =5, 6, 7, . . ., and inverting the values of m-level symbols in the odd burst pairs;c) supplementing every burst of M m-level symbols with n predetermined symbols to the total number of M+n =2
^{L }and inverting the values of the half of said predetermined symbols supplemented to the bursts of every odd burst pair;d) in each of the I and Q channels of the transmitting side, dividing every burst of M+n symbols in halves and replacing every burst of
symbols with a set of M+n time samples using the conversion from M+n frequency domain to the time domain;
e) in each of the I and Q channels of the transmitting side, converting each of said sets of M+n time samples obtained from one of said bursts of M+n symbols to the respective sequence of M+n time samples with the second clock frequency
f) in each of the I and Q channels of the transmitting side, combining both obtained sequences of M +n time samples into one sequence of the same length, for which purpose all spectrum frequencies of one of the sequences being combined are phase turned by π/2, and adding, on sample-to-sample basis, the both sequences;
g) in each of the I and Q channels of the transmitting side, filtering the combined sequence with the second clock frequency f
_{2 }in the range of 0 to f_{2}/2;h) forming a signal for transmission using said filtered sequences;
at the receiving side:
i) separating the signal being received into a signal of the I channel of the receiving side and a signal of the Q channel of the receiving side using a double quadrature conversion and digital filtration;
j) deriving, using the sum signal of the I and Q channels of the receiving side, the second clock frequency f
_{2 }and, on the basis thereof, the first clock frequency, using for deriving the second clock frequency those frequency components in the spectra of the signal of said I and Q channels of the receiving side, the frequency Ω_{i}; value of each of which is determined by the position of those from respective n additional predetermined symbols in every said burst of M+n symbols in the step c) at the transmitting side, which values are not inverted from one burst to another;k) deriving the meander signal of the interval frequency 1/T by means of comparing in phase said sum signal of the I and Q channels of the receiving side with those frequency components in the spectra of the signals of said I and Q channels of the receiving side. the frequency Ω
_{I}value of each of which is determined by the position of those from respective n additional predetermined symbols in every burst of M +n symbols at the transmitting side, which values are just inverted from one burst to another;l) forming bursts of M+n time samples in each of said I and Q channels of the receiving side using said meander signal at the intervals having a duration of
m) dividing every formed burst of M+n time samples into two sets of
symbols in each of said I and Q channels of the receiving side using a conversion from the time domain to the frequency domain;
n) in each of said I and Q channels of the receiving side, forming a set of M symbols from both said sets of
symbols by means of discarding the n additional predetermined symbols;
o) in each of said I and Q channels of the receiving side, eliminating the phase ambiguity of the formed set of M symbols using said meander signal;
p) in each of said I and Q channels of the receiving side, forming a sequence of m-level samples of all sets of M symbols, which samples running with the frequency f
_{1}/2;q) forming an output information sequence of m-level symbols with the first clock frequency f
_{1 }by means of combining said sequences of m-level samples in both I and Q channels of the receiving side.2. The method according to
_{b }=kf_{1}.3. The method according to
4. The method according to
subjecting one of two combined sequences of M+n time samples to the Hilbert transform;
delaying another one of said two sequences for a time of performing said Hilbert transform;
whereupon performing said step of adding. on the sample-to-sample basis, said converted and delayed sequences.
5. The method according to
_{2}/2, the spectra of each of the I and Q channels comprise:
the first frequency components cos Ω
_{1}t, cos Ω_{2}t, . . . cos Ω_{n/2}t, the frequency Ω_{i }value of each of which is determined by the position of those from respective n additional predetermined symbols in every said burst of M+n symbols in the step c) at the transmitting side, which values are not inverted from one burst to another; andthe second frequency components
the frequency Ω
_{i }value of each of which is determined by the position of those from respective n additional predetermined symbols in every burst of M+n symbols at the transmitting side, which values are just inverted from one burst to another.6. The method according to
7. The method according to
downconverting the signal being received to the intermediate frequency
forming the signals of the first and second channels by means of multiplying the obtained signal of the intermediate frequency by the signals sin(ω
_{IF}−Δω)t and cos(ω_{IF}−Δω)t, respectively, whereis the frequency of approximate mismatch from the intermediate frequency
digitizing the signals with the digitizing frequency equal to
2f_{I}, in each of said first and second channels;multiplying the digitized signal of said first channel by the signal cos Δωt, and the digitized signal of said second channel by the signal sin Δωt;
subjecting the signals obtained during this multiplication to the digital filtration, thus just separating the signal being received into a signal of the I channel of the receiving side and a signal of the Q channel of the receiving side.
8. The method according to
subtracting the signal of said Q channel of the receiving side from the signal of said I channel of the receiving side;
comparing in phase the obtained difference signal with said first frequency components cos Ω
_{1}t, cos Ω_{2}t, . . . cos Ω_{n/2}t;performing the step of adjusting the frequency of approximate mismatch
by the signal obtained during the step of comparing said difference signal.
9. The method according to
comparing in phase said sum signal of the I and Q channels of the receiving side with said first frequency components cos Ω
_{1}t, cos Ω_{2}t, . . . cos Ω_{n/2}t;performing the step of adjusting said digitizing frequency by the signal obtained during the step of comparing the sum signal with the first frequency components;
deriving the second clock frequency f
_{2 }from said digitizing frequency.10. The method according to
_{1}t, sin Ω_{2}t, . . . sin Ω_{n/2}t, as a result of which deriving from the signalsthe signal
from which just deriving the signal of the interval frequency 1/T.
11. The method according to
subjecting every of said burst of M+n time samples at the interval with the duration T to the direct Fourier transform, as a result of which obtaining two sets by
symbols;
forming, from every pair of obtained sets by
symbols, one set of M symbols;
multiplying every set of M symbols by the value of the signal
as a result of which just eliminating the phase ambiguity of the formed set of M symbols in each of the I and Q channels of the receiving side.
12. A system for transmitting and receiving quadrature amplitude modulation signals with the low synchronization threshold on the carrier frequency, which system comprising:
at the transmitting side:
a channel separator intended for separating the information sequence of the m-level symbols, where m =2
^{k}, k =2, 3, . . . , running at the first clock frequency f_{1}, into an I channel of the transmitting side with the even m-level symbols and a Q channel of the transmitting side with the odd m-level symbols, the running frequency of said m-level symbols in each of the I and Q channels of the transmitting side being equal to f_{1}/2;a first and second burst formers each intended for storing bursts of M m-level symbols at the interval
where 2
^{L-1 }<M <2^{L}, L =5, 6, 7, . . ., in the respective one of the I and Q channels of the transmitting side, and for supplementing every burst of M m-level symbols with n predetermined symbols to the total number of M+n =2^{L};a first and second multipliers each intended for inverting values of the m-level symbols in the odd burst pairs in the I and Q channels of transmitting side, respectively;
a third and fourth multipliers each intended for inverting values of a half of the predetermined symbols supplemented to every odd burst pair in the I and Q channels of the transmitting side, respectively;
a third and fourth burst formers each intended for separating every burst of M+n symbols in two in the respective one of the I and Q channels of the transmitting side;
a first to fourth inverse Fourier transform units intended for replacing every burst of
symbols with a set of M+n time samples using the inverse Fourier transform;
a first to fourth parallel-to-serial converters each intended for converting, in each of the I and Q channels of the transmitting side, each of the set of M+n time samples obtained from one of said bursts of M+n symbols into a correspondent sequence of M+n time samples with the second clock frequency
first and second Hilbert transform units each intended for shifting a phase by π/2 for all frequencies of the spectrum of the respective sequence from M+n time samples in the I and Q channels of the transmitting side, respectively;
a first and second delay units each intended for delaying another one of the sequences from M+n time samples in the I and Q channels of the transmitting side, respectively, for the time of processing in the respective Hilbert transform unit;
a first and second sample-to-sample adders each intended for combining, in the respective one of the I and Q channels of the transmitting side, the both sequences from M+n time samples obtained from the similarly named Hilbert transform unit and similarly named delay unit into a single sequence of the same length;
a first and second filter units each intended for filtering the combined sequence with the second clock frequency f
_{2 }in the range of 0 to f_{2}/2 in the respective one of the I and Q channels of the transmitting side;a waveform shaper for transmission intended for forming a signal for transmission from the filtered sequences in the I and Q channels of the transmitting side;
a clock-frequency discriminator intended for forming, from the first clock frequency, all clock frequencies necessary for the operation of the transmitting side units;
at the receiving side:
a quadrature conversion unit intended for separating the signal being received into digital sample sequences in the I channel of the receiving side and in the Q channel of the receiving side;
a clock-frequency discriminator intended for extracting clock frequencies using signals in the I and Q channels of the receiving side;
a first and second buffer units each intended for dividing the digital sample sequences in the respective one of the I and Q channels of the receiving side into bursts of M+n samples and for storing those bursts;
a first and second Fourier transform units each intended for performing, in the respective one of the I and Q channels of the receiving side, the direct Fourier transform on the bursts of M+n samples and for obtaining pairs of
m-level samples;
a first and second sample extractors each intended for extracting M m-level samples from each pair of
m-level samples in the respective one of the I and Q channels of the receiving side;
a first and second phase ambiguity elimination units each intended for eliminating a phase ambiguity of the respective one of the I and Q channels of the receiving side;
a first and second converters to m-level sequence each intended for forming sequences of m-level samples in the respective one of the I and Q channels of the receiving side;
a sequence combining unit intended for combining the sequences of m-level samples from the I and Q channels of the receiving side into one sequence of m-level samples running at the first clock frequency f
_{1}.13. The system according to
_{1 }to said sequence of the m-level symbols.14. The system according to
15. The system according to
_{1}.16. The system according to
a seventh and eighth multipliers each intended for multiplying an input signal by a respective quadrature component with the frequency
where
is a frequency of approximate mismatch from the intermediate frequency
a first and second filters each intended for extracting sine and cosine components of the signal being received, respectively;
a first and second analog-to-digital converters each intended for converting a respective component of the signal being received to digital samples with the double second clock frequency f
_{2};a digital quadrature demodulator intended for demodulating signals of in-phase and quadrature channels;
a first and second optimal digital filters intended for making the optimal digital filtration of the demodulated signals of the in-phase and quadrature channel, respectively.
17. The system according to
a ninth and tenth multipliers each intended for multiplying the sine component of the input signal by the respective quadrature component of the frequency
an eleventh and twelfth multipliers each intended for multiplying the cosine component of the input signal by the respective quadrature component of the frequency
a controlled frequency synthesizer forming, from the intermediate frequency adjusting signal, the sine component of the signal with the frequency
for feeding to the ninth and eleventh multipliers and the cosine component of the signal with the frequency
for feeding to the tenth and twelfth multipliers;
a subtractor intended for determining a difference of the signals from the tenth and eleventh multipliers;
a summer intended for summing the signals from the ninth and twelfth multipliers.
18. The system according to
a first to third phase-locked-loop (PLL) units, which first and second inputs are intended for receiving signal of the I and Q channels of the receiving side, respectively;
the output of the first PLL unit is intended for producing an intermediate frequency adjusting signal fed to the quadrature conversion unit;
the first and second outputs of the second PLL unit are intended, respectively, for producing a signal with the frequency
fed to the first and third PLL units, first and second buffer units, and first and second DFT units, and a meander signal
of interval frequency fed to the first and second phase ambiguity elimination units;
the output of the third PLL unit is intended for producing a clock frequency adjusting signal to a frequency component forming unit;
the frequency component forming unit, which input is intended for receiving the clock frequency adjusting signal from the third PLL unit;
the first group outputs of the frequency component forming unit are connected to respective inputs of the group of inputs of the first and third PLL units, and second group outputs are connected to respective group of inputs of the second PLL unit;
the first output of the frequency component forming unit is intended for producing signals with the frequency
to the quadrature conversion unit;
the second output of the frequency component forming unit is intended for producing signals with the second clock frequency
to the first and second buffer units;
the third output of the frequency component forming unit is intended for producing signals with the frequency kf
_{1 }to the converter to binary sequence;the fourth output of the frequency component forming unit is intended for producing signals with the first clock frequency f
_{1 }to the sequence combining unit and to the converter to binary sequence;the fifth output of the frequency component forming unit is intended for producing signals with the frequency f
_{1}/2 to the first and second converters to m-level sequence and to the sequence combining unit.19. A non-transitory computer-readable medium intended for the direct operation as a part of a computer and comprising a program for implementing the method according to
20. A method for synchronizing the reception of Quadrature modulation (QAM) signals at the interval
comprising utilizing the method of
Description This application is a Continuation of PCT application serial number PCT/RU2006/000261 filed on May 24, 2006, which in turn claims priority to Russian Patent Application No. RU 2005118509 filed on Jun. 15, 2005, both of which are incorporated herein by reference in their entirety. This invention relates to the signal transmission technique. Particularly, this invention relates to the method and system for transmitting and receiving quadrature-amplitude modulation signals (QAM) with the low synchronization threshold on the carrier frequency. In transmitting and receiving signals modulated in one or another manner, a very important characteristic is the demodulation threshold, i.e., the ratio of the signal power to the noise power (signal-to-noise ratio, SNR), at which the carrier wave of the signal being received ceases to be derived, which results in loss of the reception. The demodulation threshold depends essentially on the demodulation type employed at the transmission side, and the noiseless coding type. It is known from the theory that effectiveness of any communication system is defined by the frequency and power resources thereof, i.e., by the bandwidth occupied with the signal, and by the signal power for providing required rates for transmitting and receiving information. In general, this dependence of the rate for transmitting and receiving information upon the frequency spectrum width and signal power is defined by Shannon's equation:
where C is the information transmission and reception rate, B is the frequency bandwidth of the signal being transmitted in a communication channel, P Modern communication systems represent modern technologies constructed for specific information transmission rates. The following modulation types are used most often: -
- in the satellite communication: QPSK, 8PSK, 16QAM, 32QAM;
- in relay repeater lines: BPSK, QPSK, 8PSK, 16QAM, 32QAM, 64QAM, 128 QAM, 256QAM;
- in cable lines: QPSK, 16QAM, 64QAM, 256QAM;
- in telephony: from 16QAM to 16384QAM.
The most often used types of noiseless coding employed in modern modems are Viterbi (convolutional) coding, coding with Reed-Solomon codes, trellis code modulation (TCM), turbocoding [1], and low density parity check (LDPC) coding [3, 4]. The latter is the most effective type of noiseless coding that allows, with the loss of only 0.8-1.5 dB, to achieve the maximum information transmission rates defined by the equation (1). The Table 1 shows the LDPC coding characteristics for various coding rates and modulation types. The obstacle for implementing the achieved characteristics of the LDPC coding in the modern communication systems is too high demodulation thresholds (carrier recovery thresholds) in the existing demodulators. Thus, for the QPSK type demodulation, the existing demodulators begin the carrier synchronization at the S/N ratio of about 0 dB, for the 16QAM type modulation at the S/N ratio of about +8.9 dB, and for the 32QAM type modulation at the ratio of about +12.7 dB [2].
One can see from the Table 1 that for implementing the entire possibility of the LDPC coding for the QPSK signal at the coding rate of Ό, the demodulator should operate at the ratio The main reason of this appears from the fact that the system for carrier recovery in the modern QPSK and QAM demodulators is non-linear. There is no carrier residue in the spectrum of signals using such modulation types as QPSK, 8PSK, 16QAM, etc., therefore the wave coherent to the carrier is derived from the signal being received by means of some non-linear transformation and following filtration. But any non-linearity restricts the carrier recovery threshold. If only the carrier recovery system is linear, then the demodulation threshold would be less than −3 dB, which would permit the demodulator to keep its characteristics up to the ratio
So, the presently known noiseless coding systems, e.g., the LDPC coding and turbocoding, permit to come rather closely to the Shannon's threshold. However, its achievement is restrained by the absence of demodulators capable to operate at such low S/N ratios due to the absence of the synchronization, which requires to derive the carrier from signals utilizing such modulation types as QPSK, 8PSK, 16QAM, etc. using a non-linear transformation followed by filtration. The technique for frequency multiplying is such transformation, which technique can be implemented by raising the input signal to the M-th power (to the fourth power for the QPSK, to the eighth power for the PSK, etc.). But in doing so, a noise is raised to the same power. Moreover, a phase ambiguity emerges, too, which deletion requires for adding to the signal being transmitted a relative coding that introduces additional power loss. Complexity associated with the use of the PSK, QPSK and 8PSK modulation types is obviously demonstrated in the U.S. Pat. No. 6,697,440 (published Feb. 24, 2004) and Japan Laid-out Patent Application No. 2000-032072 (published Jan. 28, 2000). As noted preciously, the quadrature-amplitude modulation (QAM) is used amongst other modulation technique in the modern communication systems. Thus, the Japan Laid-out Patent Application No. 2001-237908 (published Aug. 31, 2001) discloses the system for deriving the QAM synchronization signal, which system providing the quasi-synchronous detection. The U.S. Pat. Nos. 6,717,462 (published Apr. 6, 2004) and 6,727,772 (published Apr. 27, 2004) disclose the methods and systems for transmitting and receiving QAM signals with carrier adjustment. However, these both patents provide only the ordinary processing of the QAM signal. The disadvantage of these analogues is the impossibility for lowering the demodulation threshold in order to come near the Shannon's threshold. The object of the present invention consists in providing such method and system for transmitting and receiving QAM signals, which permit to lower the demodulation threshold by means of providing a low synchronization threshold on the carrier frequency. In order to accomplish such a result, provided are a method and system for implementing thereof, both intended for transmitting and receiving QAM signals according to the present invention. The main principle of this invention consists in supplementing the burst of M m-level QAM symbols with the predetermined symbols, which portion does not alter from one burst to another burst, and another portion thereof is inverted periodically in some of bursts. Owing to this, at the receiving side, the QAM signal components corresponding to the predetermined symbols (which frequencies being known) are derived. According to those components, the inversion frequency is determined, which ensures the ambiguity deletion in adjusting the reception synchronization frequency. This provides the possibility to come close to the Shannon threshold. The aspects and features of the present invention are shown in detail in the appended claims. The detailed description serves for better understanding the claimed group of the inventions. The following detailed description is illustrated with drawings, in which the identical or similar elements have the same numerals. A view of the signal used in the system for transmitting and receiving QAM signals according to the present invention is shown in The system for transmitting and receiving QAM signals according to the present invention consists generally of transmitting side and receiving side connected with a communication channel. The transmitting side comprises an m-level symbol former A channel separator Each of first and second burst formers Each of the I and Q channels of the transmitting side has two multipliers. Each of the first and second multipliers Each of third and fourth burst formers A first to fourth inverse Fourier transform (IFT) units A first to fourth parallel-to-serial converters Each of first and second Hilbert transform units Each of first and second sample-to-sample adders Each of first and second filter units A waveform shaper The receiving side, which input is connected to the communication channel, comprises amplifying, filtering and intermediate-frequency down-converting means common for every receiver, which are not shown in Each of first and second buffer units Each of first and second Fourier transform (DFT) units Each of first and second sample extractors Each of first and second phase ambiguity elimination units A converter The clock-frequency discriminator The quadrature conversion unit The digital quadrature demodulator The first PLL unit The second PLL unit The third PLL unit The frequency component forming unit The method for transmitting and receiving QAM signals according to the present invention is implemented in the shown system as follows. An initial bit sequence with the frequency kf The obtained sequence of m-level symbols from the former In the I channel of the transmitting side, the even m-level symbols come to the first burst former In each of the I and Q channels, all values of the m-level symbols in the odd burst pairs are inverted, respectively, in the first and second multipliers. One half of the predetermined symbols supplemented to every odd burst pair in each of the burst formers Every burst of M m-level symbols, whether it is inverted or not, along with all n predetermined vales supplemented thereto, which one half ( Every semi-burst of The sets of M+n time samples from the outputs of the first-fourth IFT units Further, a step of turning the phase of all spectrum frequencies of one of two sequences of M+n time samples in each of the I and Q channels of the transmitting side by π/2 is provided (which step if the Hilbert transform in this case). This step is performed by the first and second Hilbert transform units Both sequences of M+n time samples obtained thereafter, with the turned phases and the delayed one, in each of the I and Q channels of the transmitting side come to the respective inputs of the first (in the I channel) and second (in the Q channel) sample-to-sample adder The obtained combined sequences in each of the I and Q channels of the transmitting side are filtered by the first and second filter units Finally, the obtained filtered signal in each of the I and Q channels of the transmitting side comes to the waveform shaper After passing the communication channel, the signal being transmitted comes to the receiving side ( These extracted components are the analogue signals that are fed, respectively, to the first and second analog-to-digital (A/D) converters In the digital quadrature demodulator The signals obtained at the outputs of the digital quadrature demodulator From the buffer units In the first and second sample extractors The obtained bursts of M m-level samples in each of the I and Q channels of the receiving side come to the first and second phase ambiguity elimination units Bursts of M m-level samples with the eliminated phase ambiguity in each of the I and Q channels of the receiving side are fed in parallel to the first and second converters The thus obtained sequences of m-level samples are combined in the sequence combining unit The converter Refer now to the operation of the clock-frequency discriminator In the first PLL unit The third PLL unit In the second PLL unit The signal from the third PLL unit The formers The signal from the second frequency former It can be appreciated by those skilled in the art that all operations of the method for transmitting and receiving QAM signals according to the present invention can be implemented not only in a hardware form but also in a software form, since the processed signal is already sampled, digitized, and transferred to the bit sample form. These samples will be processed in the computer processor in accordance with the program which algorithm was in fact described above. In this case, the program, corresponding to the implementation of the above functioning algorithm, by which program implementation in the computer it is possible to implement the method according to the present invention, could be recorded to the computer-readable medium intended for the direct operation as a part of the computer. Moreover, the method according to the present invention could be expediently applied for only synchronizing the reception of the quadrature modulation signals at the interval Therefore, all indicated possibilities are included in the form of individual aspects in the appended claims that fully defines the scope of the present invention taking into account all equivalents of the features used in this claims. The description serves only for the purpose of illustrating and explaining the principles rather than limiting the scope of the present invention. Patent Citations
Non-Patent Citations
Classifications
Legal Events
Rotate |